Sentiment Analysis for Products Review based on NLP using Lexicon-Based Approach and Roberta
Bobby Kumar, Sheetal Sheetal, Veena S Badiger, Anitha DSouza Jacintha
Abstract
Sentiment analysis is also often mentioned as opinion mining, it is a part of natural language processing technique to predict or perform research by considering the sentiment or emotional tone indicated in a written document—like a product review—is ascertained. It is used to understand whether the author of the review has a positive, negative, or neutral opinion about the product or service they are discussing. Businesses and organizations can use sentiment analysis as a useful tool to determine areas for improvement, measure consumer happiness, and make data-driven choices. This study seeks to add to the ongoing conversation in the field of NLP. It is intended to provide useful insights for academics wanting to investigate the varied field of sentiment analysis, professionals hoping to use consumer sentiment to inform strategy, and companies hoping to succeed in the fast-paced environment of the online market. The research presented here highlights the crucial part sentiment analysis plays in contemporary business and consumer decision-making processes by using Vader and Roberta. Roberta outperformed over Vader and the accuracy was around 91%.